Screening-type computed tomography (CT) of the chest, directed to the detection of small solitary pulmonary nodules (SSPN), is developing to such an extent that both its cost and radiation dose are approaching those of the traditional chest radiograph (CXR), while its sensitivity in the detection of SSPN is much higher. This development dramatically brightens the prospects for cost-effective screening for early, curable lung cancer in high-risk people. Complete evaluation of the implications of this development is a matter of quantifying 1) how often screening-type CT of the chest leads to detection of SSPN in asymptomatic persons at high risk for lung cancer. 2) how often such a SSPN - or one representing an incidental finding - is malignant, and 3) how often such a malignant SSPN is curable. Our team of investigators is committed to address all three of these questions. To address the first question, some 1000 persons at high risk for lung cancer will be screened for SSPN in the context of periodic health examinations, using both CXR and screening-type CT. The primary aim in this component is to determine the prevalence of CT-detectable SSPN as a joint function of various risk-relevant characteristics of the person. The prevalence of malignancy among the detected cases of SSPN will also be addressed among the screening-detected cases of SSPN. For complete development of the diagnostic function, a larger total series of CT-detected SSPNs will be derived from a multi-institutional SSPN "registry". A subsequent study of curability based on the multi-institutional registry, is not a part of this research application, but the protocol for this is presented as part of our overall plan. The CT images of the nodules will be analyzed in two and three dimension and relevant measures of their size, shape, density will be developed. Techniques to differentiate nodules from vessels and other abnormalities in the lung will be developed.